Machine Learning for Forensic Occupancy Detection in IoT Environments

The adoption of the Internet of Things (IoT) has brought many advantages, but it also presents challenges for the field of Digital Forensics. The heterogeneity of the data directly affects the investigative process in scenarios involving IoT applications. Through the analysis of a comprehensive and heterogeneous dataset collected from IoT devices, this study analyzes the use of machine learning algorithms to detect specific patterns to estimate the number of people in physical environments involving IoT devices. In this work, we discuss the use of Machine Learning approaches to enhance criminal investigations based on data collected from IoT environments. The experimental evaluation not only showcases the potential enhancement of Digital Forensics through the utilization of IoT data but also serves to emphasize the effectiveness of machine learning-based approaches in these environments.